Metabolite identification in rat brain microdialysates by direct infusion nanoelectrospray ionization after desalting on a ZipTip and LTQ/Orbitrap mass spectrometry John C. L. Erve 1 * , Chad E. Beyer 2 , Lawrence Manzino 2 and Rasmy E. Talaat 1 1 Wyeth Research, Drug Safety Metabolism, Collegeville, PA 19426, USA 2 Wyeth Research, Discovery Neuroscience, Princeton, NJ 08543, USA Received 29 July 2009; Revised 12 October 2009; Accepted 13 October 2009 Analyzing brain microdialysate samples by mass spectrometry is challenging due to the high salt content of the artificial cerebral spinal fluid (aCSF), low analyte concentrations and small sample volumes collected. A drug and its major metabolites can be examined in brain microdialysates by targeted approaches such as selected reaction monitoring (SRM) which provides selectivity and high sensitivity. However, this approach is not well suited for metabolite profiling in the brain which aims to determine biotransformation pathways. Identifying minor metabolites, or metabolites that arise from brain metabolism, remains a challenge and, for a drug in early discovery, identification of metabolites present in the brain can provide useful information for understanding the pharmaco- logical activity and potential toxicological liabilities of the drug. A method is described here for rapid metabolite profiling in brain microdialysates that involves sample clean-up using C18 ZipTips to remove salts followed by direct infusion nanoelectrospray with an LTQ/Orbitrap mass spectrometer using real-time internal recalibration. Full scan mass spectra acquired at high resolving power (100 K at m/z 400) were examined manually and with mass defect filtering. Metabolite identification was aided by sub-parts-per-million mass accuracy and structural characterization was accomplished by tandem mass spectrometry (MS/MS) experiments in the Orbitrap or LTQ depending on the abundance of the metabolite. Using this approach, brain microdialysate samples from rats dosed with one of four CNS drugs (imipramine, reboxetine, citalopram or trazodone) were examined for metabolites. For each drug investigated, metabolites, some of which not previously reported in rat brain, were identified and characterized. Copyright # 2009 John Wiley & Sons, Ltd. Major depression is a chronic, heterogeneous disorder affecting an estimated 21% of the global population. Several classes of antidepressants currently available to treat patients include the monoamine oxidase inhibitors, tricyclics (e.g. imipramine), selective serotonin reuptake inhibitors (e.g. citalopram, trazodone) and norepinephrine reuptake inhibi- tors (e.g. reboxetine), all of which elevate levels of mono- amines such as serotonin (5-HT), norepinephrine and/or dopamine. While all of these medicines are moderately effective in some depressed patient populations, there are still considerable limitations associated with all commercially available antidepressants such as delayed onset of efficacy, treatment resistance and deleterious side effects such as emesis and sexual dysfunction. 1 Factors that can determine how well a patient responds to a given antidepressant include absorption, distribution, metabolism and excretion of the drug. Of these factors, hepatic metabolism can differ markedly among patients and may contribute to inter-individual variability with respect to drug efficacy. Moreover, an in- depth understanding of the metabolism of a central nervous system (CNS) drug is complicated due to the potential for the brain to contribute to its metabolism. 2 Although much is known about the hepatic CYP450 enzymes which contribute to the metabolism of CNS drugs, it is now recognized that CYP450 enzymes in the brain also contribute to phase I metabolism. 2 Moreover, there is a growing awareness that CYP450 levels can vary among anatomically distinct regions in the brain and that they may display different stereospecific metabolism towards the same CNS drug than their hepatic counterparts. 3,4 Other enzymes contributing to phase I metabolism in the brain include monoamine oxidase enzymes which are also important for CNS drugs. In addition to oxidative processes, the phase II enzyme glucuronosyltransferases that catalyze formation of glucuronide metabolites are also present in brain. 5 The example of morphine-6-glucuronide which possesses more potent central acting antinociceptive properties than mor- phine 6 points to the potentially important biological activity that this class of metabolites can have in the brain. RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom. 2009; 23: 4003–4012 Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4341 *Correspondence to: J. C. L. Erve, Wyeth Research, Drug Safety Metabolism, 500 Arcola Road, Collegeville, PA 19426, USA. E-mail: john_erve@hotmail.com Copyright # 2009 John Wiley & Sons, Ltd.